The I4U mega fusion and collaboration for NIST speaker recognition evaluation 2016

K. A. Lee, V. Hautamäki, T. Kinnunen, A. Larcher, C. Zhang, A. Nautsch, T. Stafylakis, M. Rouvier, W. Rao, F. Alegre, J. Ma, M. W. Mak, A. K. Sarkar, H. Delgado, R. Saeidi, H. Aronowitz, A. Sizov, H. Sun, T. H. Nguyen, G. WangB. Ma, V. Vestman, M. Sahidullah, M. Halonen, A. Kanervisto, G. Le Lan, F. Bahmaninezhad, S. Isadskiy, C. Rathgeb, C. Busch, G. Tzimiropoulos, Q. Qian, Z. Wang, Q. Zhao, Tianzhou Wang, H. Li, J. Xue, S. Zhu, R. Jin, T. Zhao, P. M. Bousquet, M. Ajili, W. B. Kheder, D. Matrouf, Z. H. Lim, C. Xu, H. Xu, X. Xiao, E. S. Chng, B. Fauve, K. Sriskandaraja, V. Sethu, W. W. Lin, D. A.L. Thomsen, Z. H. Tan, M. Todisco, N. Evans, H. Li, J. H.L. Hansen, J. F. Bonastre, E. Ambikairajah, Gang Liu

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

11 Citations (Scopus)
160 Downloads (Pure)

Abstract

The 2016 speaker recognition evaluation (SRE'16) is the latest edition in the series of benchmarking events conducted by the National Institute of Standards and Technology (NIST). I4U is a joint entry to SRE'16 as the result from the collaboration and active exchange of information among researchers from sixteen Institutes and Universities across 4 continents. The joint submission and several of its 32 sub-systems were among top-performing systems. A lot of efforts have been devoted to two major challenges, namely, unlabeled training data and dataset shift from Switchboard-Mixer to the new Call My Net dataset. This paper summarizes the lessons learned, presents our shared view from the sixteen research groups on recent advances, major paradigm shift, and common tool chain used in speaker recognition as we have witnessed in SRE'16. More importantly, we look into the intriguing question of fusing a large ensemble of sub-systems and the potential benefit of large-scale collaboration.

Original languageEnglish
Title of host publicationProceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
PublisherInternational Speech Communication Association
Pages1328-1332
Number of pages5
Volume2017-August
ISBN (Print)978-1-5108-4876-4
DOIs
Publication statusPublished - 2017
MoE publication typeA4 Article in a conference publication
EventInterspeech - Stockholm, Sweden
Duration: 20 Aug 201724 Aug 2017
Conference number: 18
http://www.interspeech2017.org/

Publication series

NameInterspeech: Annual Conference of the International Speech Communication Association
ISSN (Electronic)1990-9772

Conference

ConferenceInterspeech
CountrySweden
CityStockholm
Period20/08/201724/08/2017
Internet address

Keywords

  • Benchmark
  • Call My Net
  • Fusion
  • Speaker recognition evaluation

Fingerprint Dive into the research topics of 'The I4U mega fusion and collaboration for NIST speaker recognition evaluation 2016'. Together they form a unique fingerprint.

Cite this